debug

A debugging procedure for finding the underlying cause of bugs, failed commands, crashes, and unexpected behavior. It checks project memory, error details, timing, and the relevant specialist guidance before changing code.

In plain words
What is it for?
Use it when investigating errors in backend code, interfaces, databases, or security-sensitive behavior, especially when logs or recent changes need to be examined.
Why use it?
It helps avoid fixing only the visible symptom and preserves the discovered cause, fix, affected files, and remaining risks for later work.

Cursor rule for Cursor

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add rules/votruongdanh/skills-agent/debug
Clone the repo
git clone --depth 1 https://github.com/VoTruongDanh/Skills-Agent

Made for: Cursor.

Per session 53 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 910 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00053 $0.00910
Opus 5 $0.00026 $0.00455
Sonnet 5 $0.00011 $0.00182
Haiku 4.5 $0.00005 $0.00091

Measured 2d ago against content hash e46680c3b149, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

debug scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.cursor/rules/debug.mdc · 79 lines

How it starts

The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Memory Protocol

START: Read .ai-memory.md from project root. Check for known bugs, past fixes, tech stack details, common error patterns, and architecture notes. END: Update .ai-memory.md using Memory Compaction Rules with: bug, root cause, fix, files touched, lessons learned, and remaining risks.

Goal

Find the real root cause, not just the first visible symptom.

Agent Routing

  • If bug is in API/server/backend → read .kiro/skills/agents/agents/backend-specialist.md and apply its knowledge
  • If bug is in UI/rendering/CSS → read .kiro/skills/agents/agents/frontend-specialist.md and apply its knowledge
  • If bug is in database/queries → read .kiro/skills/agents/agents/database-architect.md and apply its knowledge
  • If bug may be a security issue → read .kiro/skills/agents/agents/security-auditor.md and apply its knowledge
  • Default → read .kiro/skills/agents/agents/debugger.md and apply its systematic analysis

Socratic Gate

Before debugging, verify:

  1. What is the expected behavior vs actual behavior?
  2. Is there a log, stacktrace, or error message?
  3. When did this start? (recent change, always broken, intermittent?)
  4. Have we tried this before? If any answer is missing, ASK before proceeding.

Workflow

  1. Read Memory — Load .ai-memory.md for project context and past bug history.
  2. Summarize the bug, expected behavior, and actual behavior.
  3. Gather evidence from logs, stack traces, code paths, config, and recent changes.
  4. List the top hypotheses ranked by likelihood. Eliminate previously failed hypotheses immediately.
  5. Eliminate hypotheses using direct evidence.
  6. Identify the root cause and confirm with evidence before attempting any code changes.
  7. Propose an effective solution that addresses the root cause definitively. Ensure clean code standards are met and fully update all related files.
  8. Suggest how to verify the fix and prevent regressions.
  9. Quality Gate — Read .kiro/skills/_scripts/checklist.md and run cross-cutting quality checks.
  10. Update Memory — Save root cause, fix, and lessons to .ai-memory.md.

Read the full file on GitHub · 79 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 79 lines · 53 tokens per session scan A e46680c3b149

Subscribe to this mod's changes

debug is a cursor rule published in the GitHub repository VoTruongDanh/Skills-Agent (2 stars, last pushed 4mo ago), licensed MIT. It adds 53 tokens to every session and 910 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.